Insights into polyethylene biodegradative fingerprint of Pseudomonas citronellolis E5 and Rhodococcus erythropolis D4 by phenotypic and genome-based comparative analyses

Abstract

Polyethylene (PE) is the most-produced polyolefin, and consequently, it is the most widely found plastic waste worldwide. PE biodegradation is under study by applying different (micro)organisms in order to understand the biodegradative mechanism in the majority of microbes. This study aims to identify novel bacterial species with compelling metabolic potential and strategic genetic repertoires for PE biodegradation. Pseudomonas citronellolis E5 is newly isolated from solid organic waste contaminated with plastic debris, and Rhodococcus erythropolis D4 was selected for its promising potential in biodegradable plastic determined by its genetic repertoire. P. citronellolis E5 was selected for its ability to grow on PE as the only carbon and energy source. Meaningful extracellular secreted laccase activity was also characterized for D4 during growth on PE (E5 and D4 strains have a laccase activity of (2 ± 1)×10^-3 U mg^-1 and (3 ± 1)×10^-3 U mg^-1, respectively). Despite the highest level of cell numbers recorded at 7 days of growth on PE for both strains, the patterns of the metabolic products obtained and degraded during 60 days on PE were dissimilar in the two bacteria at different sampling times. However, they mainly produced metabolites belonging to carboxylic acids and alkanes with varying numbers of carbons in the aliphatic chains. Whole-genome sequence analyses of P. citronellolis E5 compared to R. erythropolis D4 and genetic determinant prediction (by gene annotation and multiple sequence alignment with reference gene products) have been performed, providing a list of 16 and 42 gene products putatively related to different metabolic steps of PE biodegradation. Altogether, these results support insights into PE biodegradation by bacteria of the Pseudomonas and Rhodococcus genera from metabolic and genetic perspectives as a base to build up novel biotechnological platforms.

Keywords: Pseudomonas citronellolis; Rhodococcus erythropolis; gene clusters; genome analysis; laccase activity; polyethylene biodegradation.

FIGURE 1

Growth levels of individual bacterial isolates (A, B, C, E5, F, H, and I) from plastic-contaminated organic waste compared to R. erythropolis D4 on 1% PE. The growth is presented as the mean value of the absorbance recorded at OD₆₀₀ ± SD.

FIGURE 2

Specific laccase activity (U mg⁻¹ of total proteins) of the supernatants (CFS) (A) and cellular extracts (B) of new isolates A, C, E5, F, and H and R. erythropolis D4. Laccase activity was measured in the presence of 2,6-DMP after growth on M9-PE in the absence (light gray) and in the presence of 100 µM CuSO₄ (gray) added during the enzymatic assay. The laccase activity is the mean of three replicates ± SD.

FIGURE 3

Phylogenetic trees. Phylogenetic analysis of P. citronellolis E5 (A) and R. erythropolis (B) based on sequence alignments with reference strains for each genus (13 Pseudomonas and 14 Rhodococcus species). The tree was constructed based on concatemer sequences of four marker genes: 16S rRNA, rpoD, ychF, and gyrB for the Pseudomonas genus and secY for the Rhodococcus genus.

FIGURE 4

Growth kinetic assay of P. citronellolis E5 (A) and R. erythropolis D4 (B) on M9-PE up to 60 days. Growth is reported as the mean value of the absorbance recorded at OD₆₀₀ (white circle) or counts of live bacterial cells expressed as CFU mL⁻¹ (gray triangle) ± SD.

FIGURE 5

Percentage of normalized peak area of product types from P. citronellolis E5 and R. erythropolis D4 oxidation of PE with respect to the control (no inoculum) up to 60 days by GC-MSD analysis. The main product species after PE oxidation were carboxylic acids (white), ketones (gray), and alkanes (black).

FIGURE 6

Number of CDSs populating each COG category by EggNOG categorization for P. citronellolis E5. The x-axis includes the abbreviations for each COG category.

FIGURE 7

Circular map of P. citronellolis E5 genome predicted with Proksee viewer (https://proksee.ca/). The first outermost ring represents the BLAST comparison of the E5 genome with the genome of R. erythropolis D4 according to the sequence identity.

FIGURE 8

Pairwise comparison of P. citronellolis E5 and R. erythropolis D4 by whole-genome average nucleotide identity (ANI). Each strain genome is linearly depicted. The red lines indicate the correspondence between similar regions of the two genomes (95% threshold), thus representing the relatedness of the genome sequence. The heat map represents the ANI value for each orthologous match.

FIGURE 9

Percentage of CDSs in each RAST subsystem category for P. citronellolis E5 and R. erythropolis D4.

FIGURE 10

Cluster tree of selected gene products (named P) from P. citronellolis E5 (A) and R. erythropolis D4 (B) genomes against RAS (named R) (listed in Supplementary Table S2). Each clade is highlighted with a different color and named according to the RAS representing each clade. The clade number is in order from the top of each tree. The numbers on the branches represent the bootstrap values calculated for the ML method from the MEGA software with 50 bootstraps. Protein name abbreviations are reported in Supplementary Tables S2, S3 for P. citronellolis E5 and R. erythropolis D4, respectively.